This is a proposal by the American Geriatrics Society and its planning committee working with Duke University Center for the Study of Aging to prepare a one week extensive workshop on Research Methodologies for Clinical Research in Geriatrics. It is proposed for physician investigator fellows or junior faculty in geriatrics who have not yet been exposed to the basic methodologies commonly used in conducting geriatrics clinical research. This conference will build upon the conference which was held at Duke in the summer of 1989. The workshop will concentrate on the "how-to" aspects of conducting such research. It will provide an introduction upon which future learning can be based, and enough background so that the physician investigator can become conversant enough with the techniques to utilize consultants appropriately. The workshop will consist of an Introductory Overview Session, ten half-day working seminars each divided into half didactic and half informal workshop and a Summation Session. In addition the opportunity for the participants to meet one-on-one with the faculty will be provided. The participants will be expected to have identified a potential research topic and attempt to apply appropriate methodologies of study design and analysis to their subject in designing a study, as the various topics unfold through the week. The topics to be included are: Getting Started - Identifying research questions, review of literature, framing a hypothesis, identifying barriers, and creating theoretical frameworks. Epidemiologic and Survey Methods - discussion of large scale epidemiology, secondary data analysis, meta analysis. Case Series Studies - descriptive studies, cross sectional studies. Experimental Designs and Controls - randomizations, special considerations for geriatric studies, nursing home populations. Longitudinal Studies organization and methodologies setting up longitudinal studies. Decision Analysis methodologies and use, cost benefit analysis, policy studies. Diagnostic Test Evaluation - how to set up evaluative studies. Data Management - methods for management of data, computer software. Data Analysis statistical analytic methods, univariate and multivariate techniques. Appropriate reading materials will be provided in advance and the faculty will be available to critique research proposals produced by the participants.